This study explores trends in skincare products for pregnant women using text mining to analyze consumer preferences for safety and functionality. Data were collected from Naver blogs, cafes, and Facebook between September 2023 and August 2024 and processed using the Textom platform. Through text mining and semantic network analysis, 50 significant keywords were selected, and centrality measurements such as degree and eigenvector centrality were applied to assess keyword importance. CONCOR analysis categorized these keywords into structural equivalence groups, revealing key consumer priorities, including ingredient safety, moisturizing properties, and product recommendations. The findings highlight the growing demand for dual-purpose skincare products suitable for both pregnant women and infants, as well as the increasing influence of consumer reviews in product selection. Academically, this research underscores the effectiveness of text mining in understanding niche consumer markets. However, the study's reliance on specific online platforms presents a limitation, as it may not fully represent broader consumer preferences. Future research should incorporate diverse data sources and quantitative analyses to further validate product efficacy and consumer satisfaction.

Abstract
1. Introduction
2. Research method
    2.1. Research Tasks
    2.2. Data Collection
    2.3. Analysis Methods
3. Results and discussion
    3.1. Frequency Analysis
    3.2. Semantic Network Analysis
    3.3. CONCOR Analysis
4. Conclusion
References

• Hyung-Bum Park(CEO, Natural Story Inc.) | 박형범 (주식회사 천연스토리, 대표이사)
• Jeong-Yeon Park(Professor, Department of Beauty Design, Wonkwang University) | 박정연 (원광대학교 뷰티디자인학부, 교수) Corresponding author